How do I A/B test Flash against Max in production?

1. Pick one request type (for example "summarise ticket").
2. Route, say, 90% to supercosmos-flash and 10% to supercosmos-max, by user so each user sees one variant.
3. Log per request: model, usage, latency, and a quality signal (user rating, task success, escalations).
4. Compare cost per successful task, not just cost per request.
5. Move the request type to Max only if the quality gain is worth the higher cost and time.

Good to know#

• Every response has an X-Request-Id; log it so a request can be found later. Why log it
• Base URL: https://api.gateway.clatterlabs.xyz/v1, with Authorization: Bearer clsk_... and X-Clatter-Account: CL-... on every call. Quick reference
• Your key's account is responsible for your end users' text under the Abuse policy. What it means for apps

Still need help?#

• General support: support@clatterlabs.com or team@clatterlabs.xyz
• Infrastructure questions, or permanent deletion of your data (your whole account or one service): infrastructure@clatterlabs.xyz

We usually reply within the first hour, and always within 24 hours. A full support team is with you.